Anxiety disorders are the most common form of pediatric psychiatric illness, affecting up to 30% and severely impairing up to 20% of all youth prior to age 18. Unfortunately, up to 50% of children remain symptomatic even with the best available treatment, making anxiety disorders a major public health problem. A major barrier to devising new treatments for anxiety disorders is that the brain pathophysiology likely starts at birth or earlier, but little is known about the earliest stages of abnormal brain development. This proposal measures brain activity and brain connectivity in neonates at high versus low risk for developing an anxiety disorder; uncovers the first steps in the neurodevelopmental pathway that results in an anxiety disorder; and provides a framework for early identification, prevention, and new treatment development. Anxiety disorders in adults are associated with increased activity in brain networks that respond to changes in the environment or `novelty', coupled with decreased activity in brain networks that regulate this novelty response. Behavioral and EEG data suggest that these processes may start in infancy. Infants with an enduring enhanced behavioral and neural (as measured by EEG) reaction to novelty are described as having the `behavioral inhibition (BI)' temperament and are at high risk for later development of an anxiety disorder. High maternal prenatal anxiety is similarly associated with both increased reactivity to novelty and increased risk for developing an anxiety disorder in offspring. The objective of this application is to identify the specific brain networks in newborn infants that are associated with this enduring enhanced response to novelty and that represent increased risk for later development of an anxiety disorder. To achieve this objective, we will use task-based fMRI to measure regional brain activity that is elicited by sudden, unexpected auditory stimuli (`oddballs') in sleeping neonates. Activity evoked by the initial oddballs represents the initial novelty response, while activity evoked by later oddballs represents the potential regulatory response. The central hypotheses are that neonates at high risk for developing an anxiety disorder (on the basis of either high maternal anxiety or high BI) demonstrate increased activity in brain networks that respond to novelty; coupled with decreased activity in regulatory networks after repeated presentation of the stimuli. We will also use resting-state fMRI to measure network connectivity, and we predict that risk for anxiety disorders will be associated with altered connectivity in networks that respond to novelty. We will test these hypotheses by recruiting pregnant mothers and obtaining MRI in offspring (n=150) within 2 weeks of birth. We will then assess the neonates and mothers at 3 additional visits over the first 2 years of life. We will assess maternal anxiety during pregnancy with questionnaires and infant temperament in the first 2 years with observational measures. Results may open new avenues for preventative measures in high-risk infants, such as repeated exposure to new stimuli; or stimulation of problematic brain networks. Such measures would have major public health impact, by preventing the most prevalent childhood psychiatric disorder.